The source says
on_error handles any error. In v3.0.4, fallback routing catches ModelProviderError and uses on_error only for retryable provider failures. Ordinary client errors such as 400, 401, and 403 responses, plus non-provider exceptions, propagate without using this fallback.error_specific_fallbacks.py
"""
Fallback Models — Error-Specific
==================================
Use FallbackConfig for error-specific fallback routing.
- on_error: tried on any error from the primary model.
- on_rate_limit: tried specifically on rate-limit (429) errors.
- on_context_overflow: tried on context-window-exceeded errors.
When a specific fallback list matches the error type, it takes
priority over the general on_error list.
"""
from agno.agent import Agent
from agno.models.anthropic import Claude
from agno.models.fallback import FallbackConfig
from agno.models.openai import OpenAIChat
# ---------------------------------------------------------------------------
# Create Agent with error-specific fallbacks
# ---------------------------------------------------------------------------
agent = Agent(
model=OpenAIChat(id="gpt-5.6-luna"),
fallback_config=FallbackConfig(
# On rate-limit errors, try these models (in order)
on_rate_limit=[
OpenAIChat(id="gpt-5.6-luna"),
Claude(id="claude-sonnet-4-20250514"),
],
# On context-window-exceeded errors, try a model with a larger window
on_context_overflow=[
Claude(id="claude-sonnet-4-20250514"),
],
# General fallback for all other errors
on_error=[
Claude(id="claude-sonnet-4-20250514"),
],
),
)
# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------
if __name__ == "__main__":
agent.print_response("What is the meaning of life?", stream=True)
Run the Example
1
Set up your virtual environment
uv venv --python 3.12
source .venv/bin/activate
uv venv --python 3.12
.venv\Scripts\activate
2
Install dependencies
uv pip install -U agno anthropic openai
3
Export your API keys
export ANTHROPIC_API_KEY="your_anthropic_api_key_here"
export OPENAI_API_KEY="your_openai_api_key_here"
$Env:ANTHROPIC_API_KEY="your_anthropic_api_key_here"
$Env:OPENAI_API_KEY="your_openai_api_key_here"
4
Run the example
Save the code above as
error_specific_fallbacks.py, then run:python error_specific_fallbacks.py